international journal of hydrogen energy xxx (xxxx) xxx

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/he

Co-gasification between coal/sawdust and coal/

wood pellet: A parametric study using response
surface methodology

Fatin Zafirah Mansur a, Che Ku Mohammad Faizal a,*,

Minhaj Uddin Monir a,b, Nur Asma Fazli Abdul Samad c,
Samson Mekbib Atnaw d, Shaharin Anwar Sulaiman e
a
Faculty of Engineering Technology, University Malaysia Pahang, 26300 Gambang, Pahang, Malaysia
b
Department of Petroleum and Mining Engineering, Jessore University of Science and Technology, Jessore, 7408,
Bangladesh
c
Faculty of Chemical and Natural Resources Engineering, University Malaysia Pahang, 26300 Gambang, Pahang,
Malaysia
d
College of Electrical and Mechanical Engineering, Addis Ababa Science and Technology University, Ethiopia
e
Department of Mechanical Engineering, Universiti Teknologi Petronas, 32610 Bandar Seri Iskandar, Perak,
Malaysia

highlights graphical abstract

Co-gasification performance on
raw biomass and pre-treated
biomass with coal.

Effect of operating variables on the
performance parameters are stud-
ied using RSM.

Experimental results are in good
agreement with the predicted
values.

The optimum H2/CO ratio and
HHVsyngas are found at high ER for
both co-gasification.

article info abstract

Article history: This study had compared raw biomass and pre-treated biomass co-gasified with coal with
Received 24 September 2019 the aim of investigating the reliability of pre-treated biomass for enhancing gasification
Received in revised form performance. Sawdust (SD) and wood pellet (palletisation form of sawdust - WP) and
2 March 2020 blends of these two feedstocks with sub-bituminous coal (CL), were gasified in an air at-
Accepted 5 April 2020 mosphere using an external heated fixed-bed downdraft gasifier system. Response surface
Available online xxx methodology (RSM) incorporating the central composite design (CCD) was applied to assist

* Corresponding author.
E-mail address: mfaizal@ump.edu.my (C.K.M. Faizal).
https://doi.org/10.1016/j.ijhydene.2020.04.029
0360-3199/© 2020 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
2 international journal of hydrogen energy xxx (xxxx) xxx

Keywords: the comparison of all operating variables. The three independent variables were investi-

Co-gasification gated within a specific range of coal blending ratios from 25% to 75%, gasification tem-

Coal perature from 650 C to 850  C and equivalence ratio from 0.20 to 0.30 against the

Sawdust dependent variables, namely the H2/CO ratio and higher heating value of the syngas

Wood pellet (HHVsyngas). The results revealed the H2/CO ratio and a higher heating value of the syngas

RSM of more than 1.585 and 6.072 MJ/Nm3, respectively. Findings also showed that the H2/CO
ratio in the syngas from CL/WP possessed a higher value than the CL/SD. In contrast, CL/SD
possessed a higher heating value for syngas with about 1% difference compared to the CL/
WP. Therefore, co-gasified coal with wood pellets could potentially be a substitute for
sawdust.
© 2020 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

respectively, compared to raw switchgrass [11]. The analysis

Introduction
Overconsumption of fossil fuel due to global energy demands
has dramatically increased over the past several decades as a
consequence of the advancement in industrial fields and
improvement in the living standards on a worldwide scale
resulting in the scarcity of energy sources [1]. Biomass is
viewed as a carbon-neutral form of energy that provides a
significant source of renewable energy [2]. The addition of a
high calorific value and presence of an appropriate catalyst
(coal) into the biomass provides a heat source for biomass
gasification, which enhances the reaction rate and stabilises
the gasification process [3,4]. Thus, utilizing biomass and coal
has been a passage between renewable and non-renewable
resource-based energy production, especially in co-
gasification [5,6]. Raw biomass with a high moisture content
and low energy density together with high transportation
costs hinders its application in the corresponding gasification
process [7,8]. On the improvement of biomass properties,
various pre-treatment methods have been developed to
enhance its implementation. Densification is a solution to
tackle low-density technical issues. Gasification of pellet fuel
is widely applied in commercial gasification resulting in a
stable syngas composition by maintaining a steady and effi-
cient gasification process. The uniform shape and density of
the pellet fuel aids in smooth feeding by making less biomass
bridges and gasification reactions [9]. Due to these merits,
pelletized biomass is frequently applied in gasification, espe-
cially in fixed-bed gasifiers, where mechanically, substantial
fuel particles of limited size are required for successful oper-
ations [10]. Thus far, only several studies have explored the
gasification performance of pelletized biomass with its parent
biomass [11,12]. Yoon conducted the downdraft fixed-bed
gasifier using rice husk and rice husk pellets and found that
rice husk pellets produced a better syngas composition of
18.6%, 20.2%, 8.1%, and 1.5% for H2, CO, CO2, and CH4,
respectively. In addition, the higher gas value and cold gas
efficiency of the rice husk pellet increased significantly (15%)
compared to raw rice husk. Densified switchgrass also resul-
ted in better gasification performance at 4.46 MJ/m3, 89.58%
and 63.98% for lower heating value (LHV) of syngas, carbon
conversion efficiency (CCE) and cold gas efficiency (CGE),
by Ref. [11] also revealed that pelletized switchgrass exhibits a
severely disintegrated fibrous structure, whereas the fibrous
matrix in the unpelletized switchgrass was still visible. These
deviations have led to efficient performances during gasifi-
cation. Co-gasification of garden waste pellets with Indian
coal shows an increase in biomass ratio resulting in the
highest production of H2 and CO with the lowest heating value
of 3.05 MJ/Nm3 [13]. Furthermore, co-gasification’s modelling
of coal with wood pellet revealed that the CO concentration
decreased and H2 increased with the increase in the blend
ratio. For the gasification performance, carbon conversion and
cold gas efficiencies for all blend ratios were more than 99.8%
and 88.1%, respectively [14]. However, to the best of our
knowledge, few conclusions have been drawn between the co-
gasification of coal with raw biomass or pre-treated biomass,
with either possessing better gasification performance with
similar parameters. Considering these factors, a comprehen-
sive investigation on the optimization of raw biomass and
pre-treated biomass for co-gasification with sub-bituminous
coal should be performed by applying the response surface
methodology (RSM) with various responses and outputs.
Response surface methodology (RSM) is an effective
statistical tool for research purposes, which employs both
statistical and mathematical methods to develop, improve
and optimize the process, as well as to investigate several
important responses affected by some variables [15].
There are many downdraft gasifiers developed worldwide
for co-gasification; unfortunately, very few studies have
been done on the prediction of the gasification perfor-
mance and incorporate process parameters, like coal
blending ratio, temperature and equivalence ratio. More-
over, there are no studies on RSM related to pre-treated
biomass, especially with coal or comparative studies on
RSM using raw biomass or pre-treated biomass with sub-
bituminous coal.
In the present study, the physical and chemical properties
of sub-bituminous coal (CL), raw biomass-sawdust (SD) and
pre-treated biomass-wood pellet (WP) were investigated to
determine the characteristic of each feedstock for corre-
sponding co-gasification investigations. Furthermore, the co-
gasification of biomass with a blend of different coal
blending ratios, gasification temperature and equivalence

Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
 international journal of hydrogen energy xxx (xxxx) xxx 3

Fig. 1 e Images of coal, CL (a), sawdust, SD (b) and wood pellet, WP (WP) that were used throughout this study.

ratios has been investigated in an externally heated down- Co-gasification experiment

draft fixed-bed gasifier system. The H2/CO ratio in the syngas
and the higher heating value of the syngas (HHVsyngas) were The co-gasification of CL/SD and CL/WP was carried out in a
obtained through a series of experimental investigations. A fixed bed downdraft gasifier located at the biomass laboratory,
central composite design (CCD) incorporating the RSM under Department of Mechanical Engineering in Universiti
method was applied to reveal the effects of interaction be- Teknologi Petronas (UTP), Perak. The schematic diagram for
tween several operating variables. The results indicate that the complete gasifier system comprising three main compo-
the models developed through the Design Expert 10 fit nents, namely the gasifier reactor, gas cooling associated with
agreeably with the experimental results for all the response the cleaning system and the online gas analyser, is shown in
variables involved. Fig. 2. The gasifier was operated at atmospheric conditions of
1e2 h for each experiment. Air, used as a gasifying agent, was
supplied to the gasifier using a 1.5 kW air compressor at a
Methodology pressure of 2 bar. Meanwhile, the rotameter, which monitors
air flow control, was attached to the gasifier system. The in-
Feedstock preparation and analysis ternal diameter and height of the gasifier were 80 mm and
500 mm, respectively. Besides, an electric furnace enclosed
The CL/SD and CL/WP blends were used as feedstock in this around the reactor mainly functioned to heat the gasifier. The
study. The CL was obtained from an electric utility company in gasifier was then flushed with compressed air at the required
Malaysia through a third-party company. Meanwhile, the SD amount of volume (0.12e0.24 Nm3/h) for 10 min before the
and WP were obtained from a wood industry company located experiment to achieve a stable state. Subsequently, the
in Penang, north of Malaysia. Both SD and WP are recognized reactor was heated to the desired gasification temperature.
as the by-product of woodworking operations and pre-treated Once the reactor achieved a stable state with the desired
sawdust in a pelletized form. Fig. 1 shows the images of CL, SD
and WP. Both CL and SD were received in powder form, while
WP was in solid form. The dimensions of the feedstock were
determined by using the Brunauer-Emmett-Teller (BET) the-
ory and the Malvern particle sizer model. The feedstock was
received prior to the day of the experiment from a local sup-
plier. The feedstock was prepared for investigating the
texture, structure and chemical properties of CL, SD and WP as
well as corresponding co-gasification experiments. Samples
were first weighed to obtain the required mass needed
throughout the experiment. Next, samples were dried in an
oven at a temperature of 105  C for about 24 h. The main
purpose of the drying process was aimed at removing most of
the moisture in the samples. Subsequently, samples were
mechanically mixed into different blending ratios (25%, 50%,
and 75%). Then, the dried samples were placed into an air-
tight container to avoid any moisture coming in contact
with the samples. The properties of the feedstocks were
determined by a proximate analysis as per the ASTME1131 [16]
procedure using a thermogravimetric analyser (Hitachi STA
7000), ultimate analysis as per the ASTM D3176 [17] procedure
using an elemental analyser (Elementar: vario MACRO cube),
and higher heating value (HHV) using a bomb calorimeter (IKA
Fig. 2 e The schematic diagram of the external heated
C200).
fixed-bed downdraft gasifier system.

Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
4 international journal of hydrogen energy xxx (xxxx) xxx

gasification temperature and air flow rate, a sample weighing
approximately 100 g was loaded into the gasifier reactor by
applying the drop-chute method. Produce gas was let into the
gas analysis unit. The gas analysis was carried out until all the
samples were fully consumed, which was for about 20 min,
and the gaseous composition displayed negative values. Once
the process was completed, the electric furnace and air
compressor was switched off and the gasifier was left to cool.
The solid residue, which was characterized as char from the
experiment, was collected from the gasifier at the bottom
opening. The final weight of the remaining char was then
weighed using a precision weight balance once it reached
room temperature. The temperature and gas composition
readings from the experiment were obtained from a desktop
computer for analysis purposes.
From the value of the gaseous product obtained from the
gas analysis unit, the H2/CO ratio in the syngas was defined as
one of the response variables. It was an essential parameter to
study as a product of the gasification that produces syngas (H2
and CO) for the chemical synthesis (e.g. methanol, liquid fuels
or for the Fischer-Tropsch process) that generates energy [18].
Moreover, the higher heating value of the syngas (HHVsyngas)
was calculated. It was a crucial output parameter that defines
the quality of syngas produced from gasification in terms of
energy content per unit volume or mass. The higher heating
value of syngas (HHVsyngas) was calculated by considering the
percentage volume of combustible gas components in the
syngas (CO, H2 and CH4) produced from the co-gasification
experiment with a specific HHV, expressed in the following
Eq. (1).


HHVsyngas ¼ ðVCO  HHVCO Þ þ ðVCH4  HHVCH4 Þ þ VH2  HHVH2 (1)
where, V, is described as the volumetric percentage of each of
the CO, CH4 and H2 obtained from online gas analyser mea-
surements where each was multiplied by its respective HHV,
and the sum of the multiplication products represent
HHVsyngas. The HHV for each gas was 12.63 MJ/Nm3 for CO,
39.82 MJ/Nm3 for CH4 and 12.74 MJ/Nm3 for H2 as per standard
value obtained from the US National Renewable Energy Lab-
oratory (NREL) [19].
Design of the experiment
An empirical modelling approach known as Response Surface
Methodology (RSM) was chosen as the experimental design
using the Design Expert software (Version 10.0.3, Stat-Ease
Inc., MN, USA). In this study, the selected factors were opti-
mized by applying the centre composite design (CCD) in RSM.
Generally, CCD is one of the conventional designs existing in
RSM. CCD incorporates two-level full or fractional factorial
designs with the addition of axial points consisting of at least
one point at the centre of the experimental range. The CCD is
also an alternative to the full factorial three-level design since
it requires lesser number of runs instead of the stipulated
corresponding outcomes [20]. Three operating variables
involved in this study were the coal blending ratio (A), gasifi-
cation temperature (B) and equivalence ratio (C). Meanwhile,
the range of variables considered were coal blending ratio
(25e75%), gasification temperature (650e850  C), and equiva-
lence ratio (0.2e0.3). Equally important is the determination of
the higher and lower boundaries for the input variables in
regards to the preliminary results and experimental setup
limitations. Meanwhile, the H2/CO ratio of syngas and the
higher heating value of syngas (HHVsyngas) were measured as
output variables. In order to evaluate the result, an ANOVA
(analysis of variance) was generated and analysed

Table 1 e Experimental design with output response variables based on the centre composite design (CCD) generated by
Design Expert.
Run Variables Response
Coal/Sawdust (CL/SD) Coal/Wood pellet (CL/WP)
A-Coal (%) B-Temp. ( C) C-ER H2/CO ratio HHVsyngas (MJ/Nm3) H2/CO ratio HHVsyngas (MJ/Nm3)
in syngas in syngas
1 50 750 0.25 1.2723 5.3290 1.1244 5.8600
2 8 750 0.25 0.8900 4.6404 1.2528 3.4806
3 50 582 0.25 0.8681 4.0090 0.8938 3.7667
4 92 750 0.25 0.8477 4.9943 0.9847 4.3032
5 25 850 0.20 0.8864 4.9088 1.4894 5.3054
6 50 750 0.25 1.1874 5.2001 1.1556 5.2806
7 50 750 0.25 1.3693 5.5638 1.1387 5.6480
8 25 650 0.20 0.7810 4.3988 0.7659 4.0170
9 75 850 0.20 0.9357 6.0724 0.7865 4.7907
10 25 850 0.30 0.9271 4.2243 1.6910 3.9759
11 50 750 0.25 1.2202 5.3655 1.1816 5.9161
12 75 650 0.30 0.7344 3.7768 1.3950 4.3527
13 75 850 0.30 1.5837 5.4202 1.4247 4.9239
14 50 750 0.25 1.3736 4.8277 1.1042 6.0160
15 50 750 0.35 0.8758 4.2833 1.4497 3.4323
16 50 918 0.25 1.3332 5.6015 1.3819 5.1468
17 75 650 0.20 0.8026 4.0831 0.8894 3.6235
18 25 650 0.30 0.3796 4.7856 1.0340 3.4405
19 50 750 0.25 1.0726 5.1665 1.0916 5.6321
20 50 750 0.15 0.5838 4.9651 0.7305 4.3937

Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
 international journal of hydrogen energy xxx (xxxx) xxx 5

Subsequently, the experimental results produced and

Table 2 e The pore size, surface area and dimensions of
calculated were fitted into the model for both co-gasifications
the CL, SD and SDP.
for each response variable. The range of the co-gasification
Sample Pore size (nm) BET (m2/g) Dimensions (mm)
performance for the co-gasification (CL/SD) was 0.380e1.584
CL 4.08 5.71 0.00747 for the H2/CO ratio in the syngas and 3.777e6.072 MJ/Nm3 for
SD 3.77 17.10 0.00362 HHVsyngas; meanwhile, the range for the co-gasification (CL/
WP 4.70 1.35 10e50
WP) was 0.731e1.691 for the H2/CO ratio in the syngas and
3.432e6.016 MJ/Nm3 for HHVsyngas. Range of the co-
gasification performance obtained from other co-
gasifications using CL and SD as fuel indicates a difference
numerically. It is capable of analysing the interaction and
of about 20% recorded from 1.6 to 4.4 MJ/Nm3 [22], which was
individual influence of independent variables on dependent
carried out in a fluidized bed gasifier. Commonly, the high
variables. Subsequently, the regression model, significance of
value of HHVsyngas in this study might be due to the applica-
terms and lack-of-fit tests were carried out to assess the
tion of the downdraft gasifier that leads to minimising the
goodness of fit, which indicates whether the proposed models
generation of tar content and eventually enhancing the gasi-
are a statistically significant. These were subjected to the
fication performance [23]. On the other hand, the range of the
probability value (p-value) with a confidence level of 95%.
H2/CO ratio in the syngas for the co-gasification of CL and WP
Meanwhile, the precision of experimental data from the
were found to be slightly similar, such as from 0.77 to 1.52 [24].
model was determined based on the value of the regression
coefficient (R2). Since the regression coefficient was at l (within
the range of 0e1) or with the value being closer to 1, it shows
Results and discussions
that the model can provide a good prediction for the output
variables [21]. Furthermore, the interaction of factors together
Characteristics of the feedstock
with their outcomes on the response is presented in the form
of three-dimensional graphs. This graphical information was
Table 2 provides the pore structure parameters as well as the
used to simultaneously interpolate the combination of in-
surface area and dimensions of CL, SD and WP, with SD hav-
teractions to intermediate points, which are difficult to study
ing the least pore size compared to CL and WP. Meanwhile, the
experimentally.
surface area of SD is higher than WP and CL. This denotes that
Table 1 presents a full design matrix generated comple-
the higher surface area is supportive of an increase in gasifi-
mentary to the CCD design, in which the experiment was
cation reactivities [25]. In addition, the powder dimensions of
conducted based on it. A total of 20 runs were designed
CL and SD were determined. The results are consistent with
comprising eight factorial points, six axial points and six
the surface area and dimension in which the smaller di-
replicates at the centre point. In this study, the experimental
mensions of the feedstock possess a higher surface area with
run was randomized to reduce the error and effect of uncon-
less pore size.
trolled factors. The experimental data were collected to
Results of the proximate analysis, ultimate analysis and
generate the empirical second-order polynomial equation
calorific analysis (as shown in Table 3), as well as SEM mi-
employed as shown in Eq. (2).
crographs, provide some important insights regarding the
X
k X
k X
k X
k chemical and structural properties of each feedstock. Based
Y ¼ b0 þ bi xi þ bii x2ii þ bij xi xj þ ε (2) on the proximate analysis of the sample, CL contained a low
i¼1 i¼1 i¼1 j¼iþ1
percentage of MC, VM, and FC as compared to SD and WP. The
where, Y is the predicted response, b0 is the intercept, bi is the proximate analysis’s value of each CL, SD and WP are in the
linear coefficients, bii is the quadratic coefficient, and bij is the range of data marked by other researchers [26,27]. It is ex-
interaction coefficients for both variables i and j. Meanwhile, ε pected that CL has the highest gasification efficiency; how-
is referred to as the random error. ever, it has a disadvantage on the latter in the form of risks,

Table 3 e Proximate, ultimate and higher heating value of coal/sawdust and coal/wood pellet.
Samples Proximate analysis (wt. %) Ultimate analysis (wt. %) HHV (MJ/kg)
MCa VMb FCc Ash C H N S Od
100% CL 8.18 39.79 33.81 18.22 52.58 5.90 1.49 1.14 38.90 20.19 ± 0.082
100% SD 11.8 68.05 19.05 1.10 44.11 5.53 2.14 2.70 45.52 17.17 ± 0.089
100% WP 9.19 79.00 10.16 1.65 44.28 6.09 1.05 0.28 48.62 17.46 ± 0.085
Blend of coal & sawdust 75CL/25SD 12.27 47.89 36.65 3.19 50.46 5.80 1.65 40.55 1.53 18.04 ± 0.206
50CL/50SD 10.92 48.23 38.41 2.44 48.35 5.71 1.82 42.21 1.92 18.75 ± 0.099
25CL/75SD 9.65 56.86 31.35 2.14 46.23 5.62 1.98 43.86 2.31 17.94 ± 0.084
Blend of coal & wood pellet 75CL/25WP 9.92 80.85 6.38 2.85 50.50 5.86 1.38 41.33 0.92 19.00 ± 0.389
50CL/50WP 9.32 79.13 8.83 2.72 48.43 5.83 1.27 43.76 0.71 18.28 ± 0.641
25CL/75WP 9.44 80.74 7.14 2.68 46.36 5.79 1.16 46.19 0.50 18.12 ± 0.112

Moisture contenta, Volatile matterb, Fixed carbonc, By difference: Od¼ (100 - (Cþ H þ N þ S)).

Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
6 international journal of hydrogen energy xxx (xxxx) xxx

the thermal pre-treatment process that applied densification.

It can also be observed that the blend of sawdust with coal is
located a bit farther from its parent material, namely SD. It is
expected that the addition of CL into sawdust will result in
changes in characteristics of the SD itself due to synergism
[31]. Besides, the result indicates that the selected biomass
transformed into a highly volatile matter (>70%) and the FC
contents acted as favourable properties for the gasification
process [29].
In addition, the morphology of the main feedstock was
investigated by employing the Scanning Electron Microscope

Fig. 3 e Ternary diagram based on the value of VM, FC, and

AC for CL, SD, WP and its blends.

such as clogging, slagging and fouling due to high amounts of

ash compared to biomass [19]. Therefore, co-gasified CL with
SD and WP might reduce this disadvantage and result in
greater gasification efficiency. In addition, as clearly expected,
MC in WP (9.19%) reduced by 11.80% (SD) due to the thermal
pre-treatment process in densification, which it was subjected
to during the manufacturing process [9]. Moreover, the VM of
the SD and WP are almost twice of the CL, as reported by Refs.
[13]. The blending of CL/WP has the highest percentage of VM
compared to the blending of CL/SD due to blending of CL with
high VM of WP [28]. According to the difference in terms of FC,
SD/CL’s blending possesses the highest FC compared to WP/
CL due to the high percentage of FC in SD. Considering the
physical and chemical properties of the feedstock, the perti-
nence of solid fuel for thermal conversion purpose can be
foreseen by locating it on a ternary diagram, which has been
recognized as a ubiquitous approach for many years [29].
Meanwhile, from the ultimate analysis, the C content in CL is
much higher (52.58%) than SD and WP, with 44.11% and
44.28%, respectively, which might result in a high composition
of carbon-monoxide in syngas [30]. Another essential point is
that WP has a much lower N and S content than CL and SD
resulting in the least possibility of NOx and SOx generated
during the downdraft gasification process. One significant
note is that the SD in this study possessed a higher S content
compared to CL, which might cause corrosion to the metallic
parts of the gasification equipment and make syngas unde-
sirable for methanol synthesis. One evident point is the
decreasing order of HHV from CL ˃ WP ˃ SD. The amount of
HHV for each CL, SD and WP were also in the range mentioned
by other researchers [14,22]. Subsequently, HHV for all the
blending fuel ratios was between 19.00 and 17.94 MJ/kg.
Furthermore, as expected, an increase in the blends of CL with
SD and WP resulted in an increase in the feedstock’s HHV
value.
The ternary diagram in Fig. 3 also presents a summary of
the blending ratio for each CL/SD and CL/WP. It shows that SD
and WP are close to each other as both come from the same Fig. 4 e SEM Image of (a), (b), (c) are CL, SD and WP,
source, while WP is in a pelletized form as it has undergone respectively at a magnification of 2000x.

Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
 international journal of hydrogen energy xxx (xxxx) xxx 7

Table 4 e Analysis of variance (ANOVA) for co-gasification of CL/SD and CL/WP.

Source Coal/Sawdust (CL/SD) Coal/Wood pellet
H2/CO ratio HHVsyngas H2/CO HHVsyngas
SS p-value SS p-value SS p-value SS p-value
Model 1.5156 0.0008 6.6468 <0.0001 1.3048 <0.0001 14.8145 <0.0001
A-Coal 0.0750 0.0691 0.1946 0.0508 0.0641 0.0009 0.3992 0.0197
B-Temp 0.4280 0.0007 2.8690 <0.0001 0.3316 <0.0001 2.5347 <0.0001
C-ER 0.0403 0.1663 0.4290 0.0081 0.5822 <0.0001 0.5498 0.0087
AB 0.0136 0.4065 1.6965 <0.0001 0.2641 <0.0001 0.0009 0.8973
AC 0.1105 0.0330 0.0546 0.2672 0.0568 0.0014 0.9579 0.0016
BC 0.1677 0.0123 0.2511 0.0304 0.0005 0.6782 0.2275 0.0629
A2 0.3017 0.0022 0.3904 0.0105 0.0012 0.5462 5.4605 <0.0001
B2 0.0579 0.1038 0.4108 0.0091 0.0035 0.3018 2.5109 <0.0001
C2 0.4754 0.0004 0.6818 0.0020 0.0004 0.7324 5.0062 <0.0001
Residual 0.1808 0.3953 0.0299 0.5196
Lack of Fit 0.1145 0.2815 0.0899 0.8972 0.0244 0.0641 0.1681 0.7812
Pure Error 0.0663 0.3054 0.0055 0.3515
R2 0.8934 0.9439 0.9776 0.9661
Adj R2 0.7975 0.8933 0.9574 0.9356
Pred R2 0.4461 0.8338 0.8504 0.8795

(SEM) model Quanta FEI 450, as shown in Fig. 4. It shows that
the SEM image of CL is composed of a small shiny solid rock-
like structure. SD illustrates a smooth particle surface; in
contrast, the particle surface of WP is rough and compact. This
is expected due to the palletisation process in the WP that has
eliminated the moisture content in the SD and making it drier
[9].
Statistical analysis of the experimental results
As mentioned earlier, ANOVA is a powerful tool for simulta-
neously investigating several variables instead of only one
variable at a time. The repetition at the centre of the design
points found in the experimental analysis aids in minimising
experimental errors [20]. The result of the ANOVA analysis is
presented in Tables 4 and 5.
Table 4 presents the ANOVA (analysis of variance) from the
co-gasification experiments that aimed to construct empirical
prediction models generated by the Design Expert. Subse-
quently, the highest order polynomial was selected for each
model in order to maximize the adjusted R2, where the terms
were significant, and the model was not aliased. Therefore,
the quadratic model was chosen for all the operating factors in
the co-gasification of CL/SD and CL/WP, which is significant
for predicting the response variables. Generally, the gasifica-
tion performance for both the SD and WP with CL were
statistically analysed to evaluate its reliability based on the
model of the co-gasification experiments using three tests,
namely the significance of terms, lack-of-fit and R-squared
tests. Table 4 presents the results of both CL/SD and CL/WP
with each of the test of significance (on factors) and in-
teractions for three responses. It was discovered that the p-
value for the whole model determines the significance of the
variables. P-value was less than 0.05, denoting that the indi-
vidual terms in the model have a significant effect on the
response, and when the p-value is larger than 0.05, the term is
insignificant. According to Table 4, the test of significance for
the effective variables and terms are not similar for different
responses. As for the co-gasification of the CL/SD, tempera-
ture was found to effect the H2/CO ratio in the syngas and the
coal blending ratio, while the equivalence ratio had influenced
the HHVsyngas. In contrast, the co-gasification of CL/WP saw all
the operating variables affecting the H2/CO ratio in the syngas
and HHVsyngas. Furthermore, the lack-of-fit test was adopted
to determine whether inconsistencies between measured and
predicted values are associated with random or systematic
errors. The p-value for lack-of-fit was less than 0.05, denoting
that there is a statistically significant lack-of-fit at the 95%
confidence level. Hence, the p-value for the term “lack of fit” in
Table 4, which is greater than 0.05, indicates that the “lack of
fit” for the whole model for the H2/CO ratio in the syngas and
HHVsyngas for the co-gasification of both CL/SD and CL/WP is

Table 5 e The empirical correlations were developed based on ANOVA for the co-gasification of CL/SD and CL/WP as a
function of actual and coded operating variables.
Samples Index Type Inter A: Coal B: Temp C: ER AB AC BC A2 B2 C2
Blend of coal & H2/CO Actual 0.83 9.51e-3 3.31e-3 1.670 1.65e-5 0.094 0.029 2.34e-4 6.401e-6 54.169
sawdust Coded 1.25 0.074 0.18 0.050 0.041 0.12 0.14 0.15 0.064 0.14
HHVsyngas Actual 12.50 0.090 0.030 59.04 1.84e-4 0.07 0.035 2.66e-4 1.71e-5 64.870
Coded 5.24 0.12 0.46 0.16 0.46 0.08 0.18 0.17 0.17 0.16
Blend of coal & H2/CO Actual 1.59 0.033 2.40e-3 0.040 7.26e-5 0.067 1.65e-3 1.45e-5 1.59e-6 1.511
wood pellet Coded 1.13 0.069 0.16 0.19 0.18 0.084 8.27e-3 9.03e-3 0.016 3.8e-3
HHVsyngas Actual 37.13 0.04 0.076 95.645 4.27e-6 0.28 0.0337 9.95e-4 4.22e-5 175.78
Coded 5.73 0.17 0.43 0.19 0.011 0.35 0.17 0.62 0.42 0.44

Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
8 international journal of hydrogen energy xxx (xxxx) xxx

insignificant. By all means, the model was a fit with the
response data collected and necessary for the next experi-
ments. Meanwhile, the fitness for all models for each response
variable for the co-gasification of both CL/SD and CL/WP was
expressed by the R2 value. The R2 for different correlations are
provided in Table 4. In brief, the R2 values are close to 1 for all
correlations tested, hence, it indicates that all the models are
of good quality for both co-gasifications. Meanwhile, the
adjusted R2 in the regression model for the co-gasification of
CL/SD and CL/WP were in excess of or close to 80%, denoting
that the whole model is regarded as satisfactory when
considering the complicated comparative analysis between
the co-gasification of SD and WP with CL in the externally
heated downdraft fixed-bed gasifier system subjected to at-
mospheric conditions.
Table 5 demonstrates the functionality of different pa-
rameters to the three operating variables with their in-
teractions based on the ANOVA analysis. The actual

Fig. 5 e 3-Dimensional response surface plots representing the combined effect of the coal blending ratio (25e75%),
gasification temperature (650e850  C), ER (0.20e0.30) each for (aec) co-gasification of CL/SD (def) co-gasification of CL/WP on
H2/CO ratio in the syngas.

Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
 international journal of hydrogen energy xxx (xxxx) xxx
coefficients aid in predicting the actual responses using the
original unit. Equally important, the coded coefficients are
suitable for comparing the relative impact of each variable
term on the responses. Table 5 shows that the major influ-
encing conditions for each dependent variable in the co-
gasification of SD and WP with CL were found to be the tem-
perature for H2/CO ratio in syngas and the equivalence ratio in
HHVsyngas. A further comprehensive study on the interaction
factor with regression coefficients will be elaborated in the
following sections.
Effect of operating variables on the H2/CO ratio in the
syngas
Fig. 5(aef) presents the effects of the three operating variables
and its interactions with the H2/CO ratio in the syngas for the
co-gasification of CL/SD and CL/WP, respectively. The relative
importance of these variables are shown in Tables 4 and 5 The
minimum and maximum values for the H2/CO ratio in syngas
for CL/SD were 0.3796 and 1.5837, respectively. In contrast, the
minimum and maximum values for the H2/CO ratio in syngas
for CL/WP were 0.7305 and 1.6910, respectively. The influence
of the operating variable for both co-gasifications on the H2/
CO ratio in the syngas was in the order of Temp.>ER >Coal.
The plot shows that the rise of the H2/CO ratio in syngas by
increasing the gasification temperature is noticeable at a
higher temperature of 850  C. Raising the temperature pro-
vides more energy, which is sufficient for breaking the
macromolecular structure of the biomass comprising poly-
mers of cellulose, hemicellulose and lignin, thus, acting as a
catalyst for improving co-gasification. Increasing the tem-
perature also leads to a decrease in CO2 and heavier hydro-
carbon contents, eventually improving the H2 formation.
Moreover, the formation of hydrocarbons and tar were con-
verted into CO and H2 by reforming reactions. Hence, it is
apparent that temperature has a significant effect on the
production of the H2/CO ratio in syngas for both co-
gasifications. When monitoring the temperature range from
650  C to 850  C for the co-gasification of CL/SD and CL/WP, the
Fig. 6 e Predicted versus actual values of (a) co-gasification of CL/SD (b) co-gasification of CL/WP on the H2/CO ratio in the
syngas.
Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
9
H2/CO ratio in syngas increased by 26% and 23%, respectively.
This is expected due to the increasing value of the H2 con-
centration and CO gaseous composition from the co-
gasification of CL/SD and CL/WP. An increase in the gasifica-
tion temperature led to the dissipation of CO2 through the
Boudouard reaction, thereby increasing the yield of CO during
co-gasification [15]. Hence, it is expected that the concentra-
tion of H2 and CO in syngas would increase during the
experiment. Based on the properties of CL/WP, it was notice-
able that blends of CL and WP possess higher VM, which
makes it a highly volatile content matter, eventually
enhancing the production of combustible gases required to
lessen the process temperature and hence, lead to a decrease
in H2 concentration [32]. Whereas in this study, co-gasification
of CL/WP eventually resulted in a high H2/CO ratio in the
syngas with highly volatile matter. This is supported by Her-
nandez et al. [33], claiming that concurrently raising the
biomass content, and temperature in co-gasification leads to a
higher production of hydrogen. In the case of the co-
gasification of CL/WP, the maximum H2/CO ratio in the syn-
gas was 25% of the coal blending ratio with the gasification
temperature at 850  C. Considering the raw and treated
(pelletized) sawdust, the WP indicates that much more effec-
tive gasification may be due to its physical aspects, such as the
large amount of pores between the SD, which makes it diffi-
cult for the heat to react efficiently by moving rapidly from the
accumulation surface to its interior.
For the operating factors of the equivalence ratio, both co-
gasifications possess similar results, whereby an increase in
the equivalence ratio from 0.20 to 0.30 causes an increase in
CL/SD and CL/WP by 8% and 29%, respectively. The increment
of H2/CO was due to the increase in the equivalence ratio
occurring at a higher temperature that resulted in higher
feedstock conversion and produced higher fuel quality [34].
This might be due to the equivalence ratio that is related to
airflow rate and residence time, which is controlled by the
degree of combustion that depends on the varying amounts of
O2 supply and leads to a significant effect on the gasification
temperature. This leads to a decrease in the energy content of
10 international journal of hydrogen energy xxx (xxxx) xxx

the gas produced because a part of the feedstock’s energy is
depleted during the combustion process in the gasification
mechanism.
Moreover, it can be seen in CL/SD that increasing the coal
blending ratio from 25% to 75% increases the H2/CO ratio in
the syngas by 12%. This is in contrast with CL/WP, in which
the value of the H2/CO ratio in the syngas had decreased by
12% as the coal blending ratio increased. This can be explained
by the low fixed carbon content of the CL/WP blends that
reduced CO production due to its consumption in the water
gas shift reaction [35] resulting in the decrease of the H2/CO
ratio as the amount of coal increases. This result was similar
to Brar et al. [36], who revealed that by increasing the amount
of coal in the hardwood pellet, the H2 and CO composition of
the product gas was reduced. In addition, it can be explained
that the presence of the oxygen constituent in the biomass

Fig. 7 e 3-Dimensional response surface plot representing combined effects for each (aec) co-gasification of CL/SD (def) and
co-gasification of CL/WP on HHVsyngas.

Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
 international journal of hydrogen energy xxx (xxxx) xxx 11

Fig. 8 e Predicted versus actual value of (a) co-gasification CL/SD (b) co-gasification CL/WP on HHVsyngas.

Predicted versus actual values of the co-gasification of CL/

Table 6 e Optimum process variables, predicted mode
SD and CL/WP are presented in Fig. 6(a) and (b), respectively.
and confirmation values of response.
The predicted and actual response values of both the co-
Parameters Limits
gasifications are shown in the quadratic regression model.
A: Coal blending ratio 25e75% CL/WP shows a higher R2 at 0.98, whereas the R2 in CL/SD is at
B: Temperature 650e850  C 0.89. The maximum value for the H2/CO ratio in CL/SD’s syn-
C: Equivalence ratio 0.20e0.30
gas was achieved with a coal blending ratio, gasification
H2/CO ratio in the syngas maximize
HHVsyngas maximize
temperature and equivalence ratio of 75%, 850  C and 0.30,
respectively. This is in contrast to CL/WP, where the
maximum values for the H2/CO ratio in the syngas comprised
25% of the coal blending ratio, gasification temperature of
results in a very high production of CO [13], eventually 850  C and 0.30 of the equivalence ratio. Meanwhile, the
reducing the H2 composition of the gaseous product leading to minimum H2/CO ratio in the syngas for both co-gasifications
a decreased value of H2/CO ratio in the syngas. of CL/SD and CL/WP was at (25%-650 C-0.30) and (50%-750

Fig. 9 e Contour plots showing coal blending ratio and temperature for each output variable at optimum condition for co-
gasification for (a) CL/SD (b) CL/WP with ER at 0.30.

Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
12 international journal of hydrogen energy xxx (xxxx) xxx
Table 7 e Optimum process variables, model prediction and confirmation values of responses.
Biomass Coal blending Temp. ( C) ER Model predicted value
ratio (%)
H2/CO
SD 75 850 0.30 1.488
WP 55 816 0.30 1.385

C-0.15), respectively. This shows that the CL/SD and CL/WP
values were placed adjacently along the linear straight line
and this denotes an agreeable regression model. Generally,
increasing the temperature and equivalence ratio results in a
high amount of H2/CO ratio in the syngas due to the co-
gasification of SD/CL. However, for the coal blending ratio,
CL/SD favours higher amounts of coal for increasing the H2/CO
ratio. Whereas, CL/WP favours a lesser amount of coal to
produce the highest amount of H2/CO ratio in the syngas.
Effect of operating variables on higher heating value of
syngas (HHVsyngas)
The effect of three operating variables on HHVsyngas for the co-
gasification of CL/SD and CL/WP are shown in Fig. 7(aef) in the
form of 3-dimensional response surface plots. Tables 3 and 4
show that both the gasification temperature and equivalence
ratio mainly affect the HHVsyngas for both SD and WP co-
gasified with CL. The maximum and minimum values of the
HHVsyngas for CL/SD are 6.0724 MJ/Nm3 and 3.7768 MJ/Nm3,
respectively. Complementary to this, the maximum and
minimum values of the HHVsyngas for CL/WP were 6.0160 MJ/
Nm3 and 3.4323 MJ/Nm3, respectively. As the gasification
temperature increased from 650  C to 850  C, both of the CL/SD
and CL/WP’s HHVsyngas increased by 17% and 15%, respec-
tively. Fig. 7 shows that the influence of the operating variable
for both co-gasifications on HHVsyngas was in a similar order as
the H2/CO ratio in the syngas. This study confirms that an
increase in the HHVsyngas is associated with higher gasification
temperatures that cause endothermic gasification reactions,
resulting in more heat loss in the system, hence, improving
the gasification process during syngas production [37].
Inversely, the HHVsyngas for the co-gasification of CL/SD
and CL/WP had decreased by about 7% with the increase of ER.
As ER is related to the airflow rate, a high ER contributes to a
higher airflow rate leading to lower heating values for syngas
and eventually reducing the efficiency of the gasification
process [3,37]. These factors contribute to the reaction’s
shorter residence time as the airflow rate increases during the
gasification process [29,38].
Moreover, the term of the coal blending ratio saw a 4% and
7% increase in the HHVsyngas as the amount of the coal
increased in the co-gasification with SD and WP, respectively.
This can be attributed to the gasification temperature and coal
blending ratio, which improves the formation of H2 and CO
that eventually increases the HHVsyngas [39,40]. Subsequently,
addition of coal in co-gasification led to a slight increase in
methane (CH4) production, eventually increasing the amount
of HHVsyngas [41].
Fig. 8(a) and (b) present the predicted versus actual values
of co-gasification of CL/SD and CL/WP, respectively. The
Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
Confirmation run value Desirability
HHVsyngas H2/CO HHVsyngas
5.535 1.58 ± 0.55 5.2683 ± 1.04 0.855
5.333 1.28 ± 0.31 4.8705 ± 0.65 0.762
predicted and actual response values of co-gasification follow
the quadratic regression model. The R2 for CL/SD and CL/WP
were calculated at 0.94 and 0.97, respectively. The maximum
value of the HHVsyngas for CL/SD are coal blending ratio,
gasification temperature and equivalence ratio with values at
75%, 850  C and 0.20, respectively. Despite this, the maximum
value of HHVsyngas for CL was at 50% of coal blending ratio, a
gasification temperature of 750  C and 0.25 of equivalence
ratio. Meanwhile, the minimum H2/CO ratio in the syngas for
the co-gasification of CL/SD and CL/WP was at (75% - 650  C -
0.30) and (50% - 750  C - 0.35), respectively. The observed
values for both CL/SD and CL/WP correspond to the predicted
value of HHVsyngas.
Process optimization and model validation
The optimization of the coal blending ratio, gasification tem-
perature, and equivalence ratio for maximum H2/CO ratio in
the syngas, HHVsyngas and Ysyngas for both co-gasifications was
performed by applying the numerical optimization method.
Table 6 presents the optimization process with a desired range
of variables and response output. The contour plots showing
optimum input variables for maximum output variables in the
co-gasification of SD and WP with coal are shown in Fig. 9.
Confirmation based on the optimized value were conducted in
triplicates. Table 7 shows the average value for each of the H2/
CO and HHVsyngas with standard deviation. A comparison
between the experimental and predicted results was calcu-
lated and shows that the error is less than 7%. It can be
concluded that the developed model can accurately predict
the output variables.
Conclusions
Extensive and comparative statistical analyses between the
co-gasification of CL/SD and CL/WP on individual and com-
bined effects of co-gasification operating variables (coal
blending ratio, temperature, and equivalence ratio) at
different process parameters of H2/CO and ratio and heating
value of the syngas (HHVsyngas) were performed by applying
response surface methodology (RSM) from a series of gasifi-
cation experiments. These experiments were carried out in
the external heated fixed-bed downdraft using air as an
oxidizing agent. The gasification temperature was found to be
a prominent variable for the co-gasification between SD and
WP. Increasing the gasification temperature resulted in an
increase of all the response variables (H2/CO ratio in the syn-
gas and HHVsyngas). Moreover, an increase in the ER in the
gasification experiment resulted in an increase in the airflow
rate, eventually minimising the residence time and in turn,
 international journal of hydrogen energy xxx (xxxx) xxx
lowering the HHVsyngas. Meanwhile, under optimum condi-
tions, results show that a maximum H2/CO ratio in the syngas
(1.58 ± 0.55) and HHVsyngas (5.2683 ± 1.04 MJ/Nm3) was ob-
tained when the coal blending ratio was at 75%, gasification
temperature at 900  C and ER at 0.30 for co-gasification of CL
with SD. Meanwhile, for the co-gasification of CL/WP, the
optimum conditions were 55% coal blending ratio, gasification
temperature of 816  C, ER at 0.30 as well as possessing a
maximum H2/CO ratio in the syngas and HHVsyngas at
(1.28 ± 0.31) and (4.8705 ± 0.65 MJ/Nm3), respectively. The
additional experiments revealed good agreement with the
predicted values. Hence, the proposed models fitted well with
the experimental results for all the response variables
involved in this study.
Acknowledgment
This study was carried out with assistance from the Biomass
Energy Laboratory (UTP) and financial support provided by the
Ministry of Higher Education, Malaysia, through a research
grant (FRGS/1/2014/TK06/UMP/02/6).
references
[1] Monir MU, Abd Aziz A, Kristanti RA, Yousuf A. Syngas
production from Co-gasification of forest residue and
charcoal in a pilot scale downdraft reactor. Waste and
Biomass Valorization 2018. https://doi.org/10.1007/s12649-
018-0513-5.
[2] Song Y, Feng J, Ji M, Ding T, Qin Y, Li W. Impact of biomass on
energy and element utilization efficiency during co-
gasification with coal. Fuel Process Technol 2013;115:42e9.
https://doi.org/10.1016/j.fuproc.2013.03.045.
[3] Valde
s CF, Chejne F, Marrugo G, Macias RJ, Go
mez CA,
Montoya JI, et al. Co-gasification of sub-bituminous coal with
palm kernel shell in fluidized bed coupled to a ceramic
industry process. Appl Therm Eng 2016;107:1201e9. https://
doi.org/10.1016/j.applthermaleng.2016.07.086.
[4] Yao Z, You S, Ge T, Wang CH. Biomass gasification for syngas
and biochar co-production: energy application and economic
evaluation. Appl Energy 2018;209:43e55. https://doi.org/
10.1016/j.apenergy.2017.10.077.
[5] Wei J, Gong Y, Guo Q, Chen X, Ding L, Yu G. A mechanism
investigation of synergy behaviour variations during blended
char co-gasification of biomass and different rank coals.
Renew Energy 2019;131:597e605. https://doi.org/10.1016/
j.renene.2018.07.075.
[6] Wei J, Guo Q, He Q, Ding L, Yoshikawa K, Yu G. Co-
gasification of bituminous coal and hydrochar derived from
municipal solid waste: reactivity and synergy. Bioresour
Technol 2017;239:482e9. https://doi.org/10.1016/
j.biortech.2017.05.014.
[7] Masnadi MS, Grace JR, Bi XT, Lim CJ, Ellis N. From fossil fuels
towards renewables: inhibitory and catalytic effects on
carbon thermochemical conversion during co-gasification of
biomass with fossil fuels. Appl Energy 2015;140:196e209.
https://doi.org/10.1016/j.apenergy.2014.12.006.
[8] Chen WH, Chen CJ, Hung CI, Shen CH, Hsu HW. A
comparison of gasification phenomena among raw biomass,
torrefied biomass and coal in an entrained-flow reactor. Appl
Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
13
Energy 2013;112:421e30. https://doi.org/10.1016/
j.apenergy.2013.01.034.
[9] Tumuluru JS, Hess JR, Boardman RD, Wright CT,
Westover TL. Formulation, pretreatment, and densification
options to improve biomass specifications for Co-firing high
percentages with coal. Ind Biotechnol 2012;8:113e32. https://
doi.org/10.1089/ind.2012.0004.
[10] Hu J, Shao J, Yang H, Lin G, Chen Y, Wang X, et al. Co-
gasification of coal and biomass: synergy, characterization
and reactivity of the residual char. Bioresour Technol
2017;5:1e7. https://doi.org/10.1017/CBO9781107415324.004.
[11] Sarkar M, Kumar A, Tumuluru JS, Patil KN, Bellmer DD.
Gasification performance of switchgrass pretreated with
torrefaction and densification. Appl Energy
2014;127:194e201. https://doi.org/10.1016/
j.apenergy.2014.04.027.
[12] Yang Z, Kumar A. The impacts of thermal pretreatments on
biomass gasification and pyrolysis processes. In:
Tumuluru JS, editor. Biomass preprocessing pretreat. Prod.
Biofuels mech. , chem. Therm. Methods. CRC Press Taylor &
Francis Group; 2018. p. 292e324.
[13] Thengane SK, Gupta A, Mahajani SM. Co-gasification of high
ash biomass and high ash coal in downdraft gasifier.
Bioresour Technol 2019:159e68. https://doi.org/10.1016/
j.biortech.2018.11.007.
[14] Jeong HJ, Hwang IS, Park SS, Hwang J. Investigation on co-
gasification of coal and biomass in Shell gasifier by using a
validated gasification model. Fuel 2017;196:371e7. https://
doi.org/10.1016/j.fuel.2017.01.103.
[15] Fermoso J, Gil MV, Arias B, Plaza MG, Pevida C, Pis JJ, et al.
Application of response surface methodology to assess the
combined effect of operating variables on high-pressure coal
gasification for H2-rich gas production. Int J Hydrogen Energy
2010;35:1191e204. https://doi.org/10.1016/
j.ijhydene.2009.11.046.
[16] ASTM E1131-98. Standard test method for compositional
analysis by thermogravimetry. West Conshohocken,PA.
1998.
[17] ASTM D3176-09. Standard practice for ultimate analysis of
coal and coke. West Conshohocken, PA: ASTM International;
2009.
[18] Wei J, Guo Q, Ding L, Yoshikawa K, Yu G. Synergy
mechanism analysis of petroleum coke and municipal solid
waste (MSW)-derived hydrochar co-gasification. Appl Energy
2017;206:1354e63. https://doi.org/10.1016/
j.apenergy.2017.10.005.
[19] Basu P. Biomass gasification and pyrolysis handbook.
Kidlington, Oxford: Academic Press; 2010. https://doi.org/
10.1016/B978-0-12-374988-8.00001-5.
[20] Karimipour S, Gerspacher R, Gupta R, Spiteri RJ. Study of
factors affecting syngas quality and their interactions in
fluidized bed gasification of lignite coal. Fuel
2013;103:308e20. https://doi.org/10.1016/j.fuel.2012.06.052.
[21] Yusup S, Khan Z, Ahmad MM, Rashidi NA. Optimization of
hydrogen production in in-situ catalytic adsorption (ICA)
steam gasification based on response surface methodology.
Biomass Bioenergy 2014;60:98e107. https://doi.org/10.1016/
j.biombioe.2013.11.007.
[22] Ve
lez JF, Chejne F, Valde

s CF, Emery EJ, London ~ o CA. Co-
gasification of Colombian coal and biomass in fluidized bed:
an experimental study. Fuel 2009;88:424e30. https://doi.org/
10.1016/j.fuel.2008.10.018.
[23] Safarian S, Unnþo
rsson R, Richter C. A review of biomass
gasification modelling. Renew Sustain Energy Rev
2019;110:378e91. https://doi.org/10.1016/j.rser.2019.05.003.
[24] Brar JS. Co-gasification of coal and Biomass. West Virginia
University; 2012.
14 international journal of hydrogen energy xxx (xxxx) xxx
[25] Wang G, Zhang J, Zhang G, Ning X, Li X, Liu Z, et al.
Experimental and kinetic studies on co-gasification of
petroleum coke and biomass char blends. Energy
2017;131:27e40. https://doi.org/10.1016/
j.energy.2017.05.023.
[26] Frau C, Ferrara F, Orsini A, Pettinau A. Characterization of
several kinds of coal and biomass for pyrolysis and
gasification. Fuel 2015;152:138e45. https://doi.org/10.1016/
j.fuel.2014.09.054.
[27] George J, Arun P, Muraleedharan C. Experimental
investigation on co-gasification of coffee husk and sawdust
in a bubbling fluidised bed gasifier. J Energy Inst 2019:2e11.
https://doi.org/10.1016/j.joei.2018.10.014.
[28] Gaqa S, Mamphweli S, Katwire D, Meyer E, Gaqa S. The
properties and suitability of various biomass/coal blends for
Co-gasification purposes. J Sustain Bioenergy Syst
2014;4:175e82. https://doi.org/10.4236/jsbs.2014.43016.
[29] Basu P. Biomass gasification and pyrolysis: practical design
and theory. Kidlington, Oxford: Academic Press; 2010.
https://doi.org/10.1016/B978-0-12-374988-8.00001-5.
[30] Brar JS, Singh K, Wang J, Kumar S. Cogasification of coal and
biomass: a review. Int J For Res 2012;2012:1e10. https://
doi.org/10.1155/2012/363058.
[31] Hu J, Shao J, Yang H, Lin G, Chen Y, Wang X, et al. Co-
gasification of coal and biomass: synergy, characterization
and reactivity of the residual char. Bioresour Technol
2017;244:1e7. https://doi.org/10.1016/j.biortech.2017.07.111.
[32] Emami Taba L, Irfan MF, Wan Daud WAM, Chakrabarti MH.
The effect of temperature on various parameters in coal,
biomass and CO-gasification: a review. Renew Sustain
Energy Rev 2012;16:5584e96. https://doi.org/10.1016/
j.rser.2012.06.015.
[33] Herna
ndez JJ, Aranda-Almansa G, Serrano C. Co-gasification
of biomass wastes and coal-coke blends in an entrained flow
Please cite this article as: Mansur FZ et al., Co-gasification between coal/sawdust and coal/wood pellet: A parametric study using
response surface methodology, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.04.029
gasifier: an experimental study. Energy Fuels
2010;24:2479e88. https://doi.org/10.1021/ef901585f.
[34] Ismail TM, El-Salam MA. Parametric studies on biomass
gasification process on updraft gasifier high temperature air
gasification. Appl Therm Eng 2017;112:1460e73. https://
doi.org/10.1016/j.applthermaleng.2016.10.026.
[35] Li K, Zhang R, Bi J. Experimental study on syngas production
by co-gasification of coal and biomass in a fluidized bed. Int J
Hydrogen Energy 2010;35:2722e6. https://doi.org/10.1016/
j.ijhydene.2009.04.046.
[36] Brar JS, Singh K, Zondlo J, Wang J. Co-gasification of coal and
hardwood pellets: a case study. Am J Biomass Bioenergy
2013;2:25e40. https://doi.org/10.7726/ajbb.2013.1005.
[37] Guo F, Dong Y, Dong L, Guo C. Effect of design and operating
parameters on the gasification process of biomass in a
downdraft fixed bed: an experimental study. Int J Hydrogen
Energy 2014;39:5625e33. https://doi.org/10.1016/
j.ijhydene.2014.01.130.
[38] Yan L, Cao Y, Li X, He B. Characterization of a dual fluidized
bed gasifier with blended biomass/coal as feedstock.
Bioresour Technol 2018;254:97e106. https://doi.org/10.1016/
j.biortech.2018.01.067.
[39] Pan YG, Velo E, Roca X, Manya  JJ, Puigjaner L. Fluidized-bed
co-gasification of residual biomass/poor coal blends for fuel
gas production. Fuel 2000;79:1317e26. https://doi.org/
10.1016/S0016-2361(99)00258-6.
[40] Monir MU, Abd Aziz A, Kristanti RA, Yousuf A. Co-
gasification of empty fruit bunch in a downdraft reactor: a
pilot scale approach. Bioresour Technol Reports
2018;1:39e49. https://doi.org/10.1016/j.biteb.2018.02.001.
[41] Alzate CA, Chejne F, Valde
s CF, Berrio A, Cruz JD La,
London ~ o CA. Co-gasification of pelletized wood residues.
Fuel 2009;88:437e45. https://doi.org/10.1016/
j.fuel.2008.10.017.